These studies are directed toward contributing to understanding the molecular basis of cellular differentiation and its relationship, if any, to malignancy. Friend erythroleukemia cells (FLC), which chronically produce Friend leukemia virus (FLV) and can give rise to malignant tumors, provide an excellent system for these studies, since they undergo erythroid differentiation both in vitro and in vivo to a hemoglobin (Hb)-producing stage. After differentiation in culture, the cells have a diminished malignant potential. Differentiation and Hb synthesis are stimulated in vitro by dimethyl sulfoxide (DMSO). 5-Bromo-2'-deoxyuridine (BUdR), a thymidine analogue, inhibits this stimulation and thymidine decreases the BUdR effect. In FLC, at least one level of control of differentiation by DMSO is at a pretranslational step, since both BUdR and thymidine act directly at transcription. Therefore, the initial investigations will focus on the mechanism of action of DMSO in stimulation of differentiation at the level of transcription. The ability of intact nuclei and cell-free chromatin prepared from both control cells and FLC, treated with DMSO and/or BUdR for various periods of time to transcribe globin mRNA and FLV RNA, will be compared by the technique of hybridization to H3-complementary DNA. Since globin mRNA has been characterized and can be highly purified, and a very specific H3-complementary DNA can be prepared from it, studies of transcription in this system should be particularly fruitful. Other proposed studies on the mechanism of DMSO action in FLC will explore possible effects of DMSO on the synthesis of proteins other than Hb, effects on transport of small molecules, the role of cyclic adenosine monophosphate in DMSO action, and the part of the structure of the DMSO molecule that affects differentiation, as well as the determination of where in the treated cell DMSO is located.